llrs
commented
2 years ago Thanks for writing this draft Joseph, sorry it took me that long to look into it. I have some comments about the file.
If the audience of the conference is for "preparing R packages for CRAN." I think that some topics are not helpful. These three topics that are more about governance than submitting or being on CRAN.
- The relation between CRAN and the R Core group
- Is there room for additional repositories?
- Should CRAN continue to accept all packages?
I agree that they are important topics and it will be beneficial to know more about them but perhaps these topics, that are the focus of this working group, should be in a special or a parallel session to be discussed.
I'm not convinced that features of a modern R repository should be for this conference, but I think it might be nice to pick up the interest of the community for new ways to do so. Last November there was a conference about software managers of programming languages, it would be probably good to follow what is currently discussed as best practices in other languages too.
In addition, I think that the most common action on CRAN has been omitted, updating a package. Although it could be seen as part of two other ideas (responsibilities of maintainers and testing packages before submitting). But I think it deserves its own point as is the most common action, and when friction usually happens.
It would also be worth to hear about packages are removed or archived from CRAN, be it copyright problems, API changes, development continues in another package, friction with tests or forbidden by the CRAN team. Hearing some good and bad histories would probably help (and probably the later more than the former).
CRAN Conference .md file in the Documents folder is for collecting ideas about the WG could hold a conference devoted to CRAN and CRAN best practices. Please add you ideas and comments.